Anionic and neutral spiked triangle clusters of mercury: a comparative study of the metal ligand redistribution process

Article abstract of DOI:10.1016/0022-328X(90)85514-Y

The action of the NEt4⁺ salts of the anions ^2- and ^- on ClHg-m complexes (m = Mo(CO)3Cp, W(CO)3Cp, Mn(CO)5, Co(CO)4, and Fe(CO)2Cp) has been investigated.In the first case, anionic spiked triangle clusters of formula ^- were obtained in high yields, and no metal ligand redistribution reactions were detected.However, the action of ^- on the same bimetallic transition metal-mercury compounds gave the neutral species , which spontaneously redistributed to 2> and m₂Hg.Only the molybdenum derivative could be isolated in pure form.The different behaviour of the two series of complexes in the redistribution process is discussed.

Full text of DOI:10.1016/0022-328X(90)85514-Y

285
Journal of Organometallic
398 (1990) 285-291
Elsevier Sequoia S.A.,
JOM 21195
Anionic and neutral spiked triangle clusters of mercury: a
comparative study of the metal ligand redistribution process
and Miquel
Departament de
Universitat de Barcelona, Diagonal 647, 08028 Barcelona (Spain)
(Received May
1990)
The action of the
salts of the anions
complexes (m =
has been investigated. In the first case, anionic spiked
were obtained in high
and
on
and
triangle clusters of formula
yields, and no metal ligand redistribution reactions were detected. However, the
action of on the same bimetallic transition
mercury compounds gave the neutral species
which spontaneously redistributed to
and
Only
the molybdenum derivative could be isolated in pure form. The different behaviour
of the two series of complexes in the redistribution process is discussed.
Although the redistribution chemistry of organomercurial compounds has been
extensively investigated, analogous reactivity for heterometallic complexes of
mercury is less well documented
In general, it is observed that most of the
two-center two-electron bonded transition-metal mercury compounds redistribute to
unsymmetric species
e.g.:
+
In contrast, with complexes containing a three-center two-electron mercury-transi-
tion metal bond formation of the symmetric species is favored
e.g.:
+
These observations can be readily accounted for in terms of kinetic factors if it is
assumed that the preferred pathway for the redistribution reactions involves a
bimolecular, associative process. This process, believed to proceed through a
1990 Elsevier Sequoia S.A.

286
center bridged transition state
would involve an increase in the coordination
number of the mercury atom, which is not unreasonable for complexes containing
two- or three-coordinate mercury atoms. However such a process is unlikely for the
mercury in the Spiro-type compound in Eq. 2 because it is already four-coordinate,
and coordination numbers greater than four are not normally observed for mercury
in transition metal complexes.
The assumption that redistribution reactions occur via an associative mechanism
prompted us to attempt the synthesis of the anionic spiked triangle clusters with a
three-center two-electron-transition metal bond of formula
where m is a metal fragment. It was assumed that the negative charge on these
derivatives would preclude unwanted symmetrization processes and facilitate the
isolation of their salts. In fact, to our knowledge, the only mercury complex with
such a geometry so far described is
to disproportionate readily to
which tends
and
The synthesis of such derivatives was accomplished by using
reactions between anionic and
condensation
metal complexes, a well known route to such
mixed-metal clusters. In addition, we describe below the synthesis and the ligand
redistribution of the structurally close neutral tetrametallic derivatives of the type
in order to make comparisons between the
behaviour of the two series of compounds in metal ligand redistribution reactions.
Results and discussion
(a) Synthesis of the anionic complexes
displaced a chlorine atom from
(m =
to afford the mixed metal clusters
+
3)
(m =
They were isolated as their
crystalline, and air-unstable solids. They can be manipulated in the air only for a
few minutes. In THF solutions are stable at -20°C for at least 24 h, after
(3);
(4);
salts in good yields; all of them are orange,
which gradual decomposition takes place, giving unidentified products. The instabil-
ity of these compounds prevented isolation of single crystals for X-ray structure
determination. The complexes have been characterized by elemental analyses as well
as infrared,
NMR and
spectroscopy (Tab. 1). It should be noted that
the halide displacement in
3 gives a product containing an edge-bridged linkage
with a three-center two-electron mercury-transition metal bond, rather than a
terminal mercury-cluster linkage with a two-center two-electron bond. A reasonable
explanation of this behaviour lies in the fact that in a large number of polynuclear
compounds the negative charge is thought to be delocalized over two or more metal

287
Table 1
Analytical and IR spectra data of complexes
NMR
(ppm)
Analyses
C
Compounds
(cm-‘)
N
H
1.36
5.39
59.20
3.68
1
(3.69)
(60.50)
1960s 1940s 189Om
1875m 1770m
1975vs
1.28
5.47
29.19
2.70
(2.50)
(28.82)
1865m 1770m
2010m
2.39
1.68
29.23
(29.26)
(2.32)
(1.62)
3.35
(3.09)
2.29
(1.97)
29.68
(29.10)
2025sh 2010m
1930s
1880m 1770m
1970s
1.72
5.37
32.31
3.18
(2.96)
(1.66)
(32.73)
1930s
1770m
centers, thus enabling the formation of compounds with a mercury coordination
number greater than two. Several trinuclear ruthenium clusters show a
behaviour when they are treated with mercury halides
Shriver has recently obtained the mixed metal cluster
and by this method
(PPN =
+))
Interestingly,
addition of a further equivalent of
to solutions of l-5 does not produce
any change, and the reagents are recovered unaltered. Steric effects probably give
rise to the inertness of these mercury clusters.
Formation of the anionic species can be monitored by IR spectroscopy, which
shows a shift for the strongest CO stretching band of about 100 cm-’ towards
higher frequency relative to the band for the starting material,
In
addition to this, compounds l-5 show absorptions in the CO bridging group region.
The
IR pattern for l-5 is very close to that shown by the similar complex
(Fig. which has been structurally char-
This unambiguously confirms the structure proposed for l-5. These
acter&d
results are not surprising bearing in mind the isolobal analogy between
and The spectrum of 1 also confirms the proposed
structure. Thus, the spectrum shows the expected quadrupole doublet (0.42 mm/s)
and the isomer shift (0.11 mm/s) at 80 K corresponds approximately to a
charge, as found for the closely related compounds
(E = Zn, Cd, and Hg)
1
Remarkably, in no case ligand redistribution could be observed. The final
products for such a process would be the compound
and the trimetal complexes Although the first of these has not yet been
obtained, the IR spectra of the latter compounds are well known, and thus their
formation in solution would have been easily detected. The IR spectrum of the
decomposed solutions even after standing for several hours at room temperature
showed no trace of

288
1800 cm- '
2000
Fig. 1. Infrared spectrum in the
(a), and
region (THF solution) of
(b). The bands marked with A correspond to the
fragment.
In order to make comparisons relative to the ease of ligand redistribution
reactions between this type of derivatives and analogous neutral clusters we under-
took the synthesis of complexes of formula
(b) Synthesis of the neutral
The reaction of
with
to give
in THF solution at 20 C was monitored by IR and
4).
NMR spectroscopy
+
(eq.
( m =
The IR pattern in the
region of compounds 6-10 consists of a number of
bands between 2040 and 1900 cm-’ (terminal carbonyl ligands) and one band at
1770 cm-’ (bridging carbonyl group). The close similarity between these spectra
and those displayed by the recently obtained complexes
(M = Cu, Ag, and Au)
enables us to propose identical
geometries for the two compounds and, on the basis of the l&electron rule, a Fe-Fe
single bond is required for these compounds to be diamagnetic, electron-precise
species. It should be noted that the mixed metal cluster formation from

is accompanied by a notable shift toward higher
frequeny of the band due to the carbonyl bridging group. THF solutions at 50
NMR spectrum consisting of a signal at
along with another invariable at 10
C
of derivatives
showed a simple
about -6 ppm (relative to
ppm, this being particularly intense for m =
Remarkably, when the temperature was allowed to raise to
and
C both signals
disappeared with concomitant formation of other peaks at 28 and 46 ppm belonging
to unidentified products. These results are interpreted in terms of a process
involving the formation of the neutral clusters
which spontaneously redistribute to give the
complex
(signal at + 10 ppm) and
(Eq. 5); the latter compounds were easily
detected by IR spectroscopy.
2
I
m
7 - 1 0
In order to confirm this hypothesis we attempted the synthesis of
by reacting
at low temperatures. The signal at
with an excess of
ppm did, indeed, appear,
along with the signal at 13 ppm corresponding to the unreacted phosphido anion.
An increase in the temperature led invariably to its rapid decomposition (signals at
28 and 46 ppm). As indicated, the redistribution process is less extensive in the case
of the molybdenum complex. These results agree with those recorded for bimetallic
complexes involving metal-mercury bonds
although this different behaviour
was not clearly understood. The greater stability of 6 towards the ligand redistribu-
tion allowed us to isolate it in a pure form in low yield (about 10%) after
chromatographic purification.
In conclusion, we have shown that the anionic heterometallic mercury species
can be isolated as their
salts in high yields,
whereas the neutral derivatives undergo spontaneous metal ligand redistribution
reactions. These facts support the idea that these redistribution reactions involve a
bimolecular, associative process.
Solvents were dried by standard methods and all manipulations and reactions
were performed in Schlenk-type flasks under nitrogen. Elemental analyses for C, H,
and N were carried out at the Institut de
de Barcelona.
spectrometer;
NMR and
shifts are
NMR spectra were recorded on a Bruker WP
relative to
and
shifts are relative to
Infrared spectra were
spectrum was
recorded on a Perk&Elmer 1330 spectrophotometer. The
recorded using a
of
source in a Rh matrix, and the calibration was with
iron foil. Compounds

and
were prepared by proce-
dures described previously.
Synthesis of
and
(m
Details of the synthesis of 1 apply also to 2-5.
A solution of (0.42 g, 0.85 mmol) in THF (25 ml) was added
to a vigorously stirred suspension of (0.51 g, 0.85
mmol) in THF (40 ml) at 20 C; the mixture turned dark red immediately. After
30 min stirring, was filtered off and the solvent was reduced to
of cold diethyl ether were then added to the filtrate. Red microcrystals were formed
when the solution was kept at 30” C for at least 12 h. After filtration, the
microcrystals were washed with a small amount of diethyl ether and dried under
vacuum. Yield:
(80%).
of
This complex was prepared by a modification of the published method
To a
was
slurry of 2.90 g of
added 0.82 ml of
(5
in 150 ml of THF at
(5 mmol). The solid
dissolved slowly
and a red solution was obtained in a few minutes. After 30 min stirring, additional
(0.2 ml) was added, and the stirring was maintained for another 10 min.
After filtration the solution was concentrated to dryness and the solid was treated
twice with 30 ml of hexane to remove the excess of phosphine. The resulting red
residue was dissolved in
ml) and diethyl ether (20 ml) was added. The
red-orange crystals obtained were filtered off, washed with diethyl ether, and dried.
Yield:
S
(50%). IR(THF)
NMR: 13 ppm.
2010 m, 1960 vs, 1930 s, 1915 vs, 1710 cm-‘.
Synthesis of
A solution of
0.67 mmol) in THF (20 ml) was added
0.67
to a stirred solution of
20 “C. The solution was allowed to warm to room
mmol) in THF (30 ml) at
temperature and then filtered, and the solvent was removed under vacuum. The
dark red residue was dissolved in the minimum quantity of THF and adsorbed onto
silica. The silica was pumped dry and applied to the top of a chromatography
column. Elution with THF-hexane (1
gave
as a red band, and the product was purified by recrystallization
from diethyl ether. Yield: 0.07 g (10%). IR (THF)
cm-‘): 2040 m, 2020 m,
cm-‘. Anal. Found: C, 34.17; H, 1.5.
NMR: -6.7 ppm; = 270 Hz.
I970 vs, 1955 s, 1905 m, 1885 s,
C, 34.53; H, 1.6. S
Acknowledgements
Financial support from the
General de
Cientifica y
(Project
is gratefully acknowledged. The authors thank A.
spectrum.
Labarta and J.R. Alabart for recording the

291
References
1 See for example, D.A. Roberts and G.L. Geoffroy, in G. Wilkinson, F.G.A. Stone, and E.W. Abel
(Eds.), Comprehensive Organometalhc Chemistry, Pergamon, Oxford, 1982, ch. 42.
2 (a) M.J. Mays and
Sweeney, J. Chem.
Organomet. Chem., 273 (1984) 233; (d) 0. Rossell, M.
Trans., (1986) 1011; (e) P. Braunstein, 0. Rossell, M.
Robb, J. Chem.
Dalton Trans., (1979) 767; (c) 0. Rossell, M.
A, (1968) 329; (b) F.
V.B.
and P. Braunstein, J.
and I. Torra, J. Chem. Dalton
I. Torra, X. Solans, and C.
0. Rossell, M. and P. Braunstein, J. Chem.
and J.J.
Organometallics,
(1986) 1113;
J.
Dalton Trans., (1987) 119.
3 (a)
Tiripicchio
K. King, K.I. Hardcastle, E. Rosenberg, A. Manotti Lanfredi, A. Tiripicehio, and M.
Inorg. Chem., 22 (1983) 1339; (b) J.A. Iggo and M. Mays, J. Chem.
Dalton Trans., (1984) 643; (c) E. Rosenberg, D. Ryckman, I-Nan Hsu, and R.W.
Chem., 25 (1986) 194; (d) T.J. Henly and J.R. Organometallics, 8 (1989) 2729.
Inorg.
4 M.H. Abraham in C.H. Bamford and C.F.H. Tipper (Eds.), Comprehensive Chemical Kinetics,
Amsterdam, 1973, Vol. 12.
5 C.P.
6 0. Rossell, M.
7 J.W. Lauher and K. Wald, J. Am. Chem.
8 S. Alvarez, M. Ferrer, R.
291.
E.M.
C.P.
and P.G. Jones, Inorg. Chem., 29 (1990) 348.
103 (1981) 7648.
and X. Solans, J. Organomet. Chem. 377 (1989)
and D.F.
J. Am. Chem.
107 (1985) 8136.
0. Rossell, M.
9 M. Ferrer, R.
10 K.
11 R.T. Baker,
0. Rossell, M.
R. Greatrex, and N.N. Greenwood, J. Chem.
Krusic, J.C. Calabrese, and D.C. Roe, Organometallics, 5 (1986) 1506.
and X. Solans, to be published.
M.
A,
2339.